Stream: dotnet
Topic: Is there FHIR validator for .NET
Yunwei Wang (May 02 2017 at 13:29):
This page https://www.hl7.org/fhir/validation.html has a JAR validator. Is there a .NET implementation?
Grahame Grieve (May 02 2017 at 22:23):
yes there is, though it is still work in progress
Brian Postlethwaite (May 03 2017 at 03:17):
It is actually a part of the current HL7.Fhir.Specification packages!
This is what is used by the new Forge editor too.
Details descriptions of its usage are not readily available yet, maybe we'll get to this at the connectathon.
(there are unit tests for it, so you might be able to work it out from those)
Brian Postlethwaite (May 03 2017 at 03:17):
(In both the DSTU2 and STU3 versions)
Richard Kavanagh (May 03 2017 at 08:25):
I can confirm it exists as we commissioned FHIR to build it for us. Furore do have a simple test harness for it, though I am not sure if that has been released anywhere. I created a different test harness here https://github.com/clarotech/Verifier/tree/develop which shows at a high level how the API can be used.
Yunwei Wang (May 03 2017 at 18:35):
I was trying to figure out how the validator validate coding using external terminology service. I am trying to make our terminology service available for that validator. I read source codes from Github. Now I understand how ExternalTerminologyService class works. It is interesting that the java implementation sends valueset as a post parameter while dotnet implementation adds valueset Id in route.
Grahame Grieve (May 04 2017 at 00:17):
the java validator doesn't assume that the terminology server has the right set of value sets
Ewout Kramer (Jun 27 2017 at 13:42):
I'd like to change the behaviour of the term validation so it will try the in-memory terminology server first, but when that fails (either because the VS is too complex or unknown), it will try an external server. Would it make sense in that case to send the valueset as a whole to the external server? It seems like a waste of bandwith.... but you cannot assume that server has your VS either (unless you manage that VS yourself). Should that be configurable behaviour?
Peter Jordan (Jun 28 2017 at 02:07):
Assuming that the purpose here is to "check that codes/displays provided in the Coding/CodeableConcept types are valid", and concerns such as realm-specific valuesets, languages, code system versions, etc. come into play, my feeling is that there needs to be a configurable option to use an external Terminology Service.
Grahame Grieve (Jun 29 2017 at 11:51):
what I do in my validator is break down the part I can't process, and pass just that to the server. It's not yet been big
Brian Postlethwaite (Jul 02 2017 at 10:41):
In my server, I just pass the reference to my internal implementation, then fall back to the memory one if don't have it loaded in my database
Last updated: Apr 12 2022 at 19:14 UTC
